Good Afternoon Everyone,
One of our customers is using Red Hat Linux version 2.4.18-14 #1 Wed Sep 4 13:35:50 EDT 2002 i686 i686 i386 GNU/Linux as a database server.
Version of Progress is 9.1D on LINUX.
Server structure was migrated to VMWare virtual machines two years ago.
Recently we are having problems with two database servers on two seperate sites.
We cannot pinpoint the exact reason.
In the middle of the production day, progress database stops responding to connection attempts. Existing connections are working ok until disconnected.
When we try to connect using promon utility it hangs.
Proshut hangs as well.
The only way of restoring the working state is to restart the server.
It works OK for a period of time before same thing occurrs again.
Nothing from a Progress perspective has changed on the servers and I can therefore only assume something has changed on the client side.
The customer uses Crystal Reporting connected to the databases in question (poor reporting structures perhaps ??)
What we do notice is when we analyse "top" linux command, several "_sqlsrv" which are using heavy processing when the issue is reported.
Is there advice on how to resolve this issue ?
Try to isolate problem by stopping all Crystal Reporting for a day ?
Move db to a different server ?Load balancing across disks ?
DB Server parameters ?
I am attaching a sample database log.
Best regards,
Brendan
One of our customers is using Red Hat Linux version 2.4.18-14 #1 Wed Sep 4 13:35:50 EDT 2002 i686 i686 i386 GNU/Linux as a database server.
Version of Progress is 9.1D on LINUX.
Server structure was migrated to VMWare virtual machines two years ago.
Recently we are having problems with two database servers on two seperate sites.
We cannot pinpoint the exact reason.
In the middle of the production day, progress database stops responding to connection attempts. Existing connections are working ok until disconnected.
When we try to connect using promon utility it hangs.
Proshut hangs as well.
The only way of restoring the working state is to restart the server.
It works OK for a period of time before same thing occurrs again.
Nothing from a Progress perspective has changed on the servers and I can therefore only assume something has changed on the client side.
The customer uses Crystal Reporting connected to the databases in question (poor reporting structures perhaps ??)
What we do notice is when we analyse "top" linux command, several "_sqlsrv" which are using heavy processing when the issue is reported.
Is there advice on how to resolve this issue ?
Try to isolate problem by stopping all Crystal Reporting for a day ?
Move db to a different server ?Load balancing across disks ?
DB Server parameters ?
I am attaching a sample database log.
Best regards,
Brendan
